Baseline variables with consistent directionality of effect on High responder status across multiple antibody outcomes. a) Risk difference, P value (Wald-test based on TMLE standard errors), and false-discovery-rate-adjusted P values for the eight baseline variables with consistent effect directionality across at least two immune response outcomes. Blank cells indicate inconsistency of effect directionality across the 10 folds of cross-validation datasets for a given antibody outcome. b) Cross-classification of gp140 High responder status (High, Non-High) x (Assigned female sex at birth, Assigned male sex at birth). c, d) Violin plots of study-pooled distributions of c) height (cm) and d) total white blood cell count (/nl) displayed according to gp140 IgG binding antibody High or Non-High responder status. The top and bottom of each boxplot within the violins indicate the interquartile range; the horizontal line is the median. gp120 IgG: n = 752; gp140 IgG: n = 772; gp41 IgG: n = 628; CD4: n = 808; gp140 IgG + gp120 IgG: n = 751; gp140 IgG or gp120 IgG: n = 751. MCV, mean corpuscular volume. WBC, white blood cell.